MCP-GROQ

Overview

MCP-GROQ implements a microservices architecture using MCP to deliver specialized AI tools. The project includes three main tools: DuckDuckGo Search Engine for web search, Mathematical Operations Engine for arithmetic calculations, and Tech News Aggregator for retrieving technology news articles. Each tool operates through a client-server architecture leveraging large language models through Groq's API.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• duckduckgo_search

  Web search functionality via DuckDuckGo with content extraction

  Try:Search for latest AI newsTry:Find information about climate change
• math_calculation

  Basic arithmetic operations with natural language processing

  Try:What is 15 multiplied by 7?Try:Calculate the square root of 144
• tech_news_retrieval

  Latest articles from Ars Technica with parsing and summarization

  Try:Get today's top tech newsTry:Find articles about artificial intelligence

Comparable tools

Installation

curl -sSf https://install.ultraviolet.rs | sh
# or with pip
pip install uv

git clone https://github.com/rahulsamant37/mcp-tools.git
cd mcp-tools
uv venv
source .venv/bin/activate  # On macOS/Linux
uv pip sync

GROQ_API_KEY=your_groq_api_key_here

uv pip install mcp

{
  "mcpServers": {
    "groq-tools": {
      "command": "python",
      "args": ["-m", "mcp.server.groq-tools"],
      "env": {
        "GROQ_API_KEY": "your-api-key"
      }
    }
  }
}

FAQ

How do I test individual tools?

Each tool has its own client script. Run `python duckduckgo_client.py` for search, `python math_client.py` for calculations, and `python weather_client.py` for news.

What is the architecture of this project?

The project implements a microservices architecture with server layer for functionality, client layer for connections, and LLM integration using Groq's Qwen-2.5-32b model with ReAct methodology.

Compare MCP-GROQ with